In this competing renewal (Year 11) of our R01 using fMRI to study late-life depression (LLD) pharmacotherapy (R01MH076079), the primary aim is to characterize functional connectivity changes associated with initial medication exposure (12-hour challenge). Our preliminary data suggests that these initial fMRI changes reflect monoaminergic engagement, regardless of monoaminergic class (serotonergic or noradrenergic), and predict later treatment response. In the proposed study we test a neural systems level model that response in LLD is mediated by acute pharmacologically-induced changes in cognitive and affective large scale network. Depression in older adults is frequently disabling and is often resistant to first-line treatments, requiring more prolonged treatment trials than in younger adults, mainly due to its heterogeneous pathophysiology (e.g. vascular and degenerative brain changes). Currently, there is little neurobiological data to guide changing or augmenting antidepressant medications. Thus, there has been a heightened focus on tailoring treatment to optimize outcome as described in the 2015 NIMH draft strategic plan (strategy 3.2). While antidepressant clinical response may take up to 8 weeks, recent studies suggest that physiologic changes, as measured with pharmacologic fMRI (phMRI) are seen within 24 hours of starting a new monoaminergic antidepressant1. For this proposal, we focus on three major Cognitive and Affective Networks (CAN): the Default Mode Network (DMN), the Salience Network (SN) and the Executive Control Network (ECN). The proposed model suggests that monoaminergic engagement leads to core CAN changes, changes that subsequently are related to overall clinical response as well as response in specific symptom clusters such as negative bias, somatizations/ anxiety and cognitive control. The same networks that are functionally connected while individuals are at rest, are also selectively engaged during tasks. Our prior work shows that pharmacotherapy ? regardless of type of antidepressant used - engages these specific networks at rest and during standard cognitive and affective tasks. Given the role of cerebrovascular disease in LLD treatment response, we will also explore the moderating role of vascular burden on the proposed association between CAN engagement and treatment response. We will recruit 100 older adults with LLD who will be randomized to receive treatment with either a very specific serotonin reuptake inhibitor (escitalopram) or a norepinephrine reuptake inhibitor (levomilnacipram). A pair of fMRI scans one day apart will be used to measure FC associated with medication titration. We propose to use a very early (12 hours after initiation of treatment) biomarker of treatment response, which, when validated, would decrease substantially the waiting time between medication changes. Additionally, our study will further our understanding of the acute neural system changes associated with monoaminergic antidepressants; this knowledge of mechanism is essential for both guiding LLD treatment research, and serving as an engagement target in LLD treatment research.